Study on the decay behavior of the stability of asphalt mixture under different dry-wet cycle conditions

Abstract

Asphalt mixtures will inevitably be affected by rainwater and the effect the of dry-wet cycle during the process of use. To explore the decay of the stability of asphalt mixtures under different dry-wet cycle conditions, a dry-wet cycle laboratory test was carried out based on standard Marshall specimens. The asphalt mixture specimens were subjected to dry-wet cycle treatment for specified cycle durations and numbers of cycles at three temperatures, and Marshall stability tests were then carried out. In this way, the decay of the stability of the asphalt mixture under different dry-wet cycle conditions was evaluated. The results show that the number of dry-wet cycles exhibited good linear relationships with the stability and the stability decay rate, respectively. As the temperature of the dry-wet cycle increased, the stability of the asphalt mixture was found to decrease; the higher the temperature, the greater the stability decay rate. A power function relationship was found between the dry-wet cycle duration and the stability decay rate of the asphalt mixture, and exhibited a decreasing trend. Under the premise of a fixed total dry-wet cycle duration, the dry-wet cycle conditions of a short duration and many cycles were found to have the greatest influence on the stability decay rate of the asphalt mixture. Moreover, a BP neural network prediction model was established by MATLAB software. By comparing the predicted and true values, the established BP neural network prediction model was found to well predict the stability of the asphalt mixture under different dry-wet cycle conditions.